Negotiating your offer

Three to six conversations, lasting maybe an hour total, will set your compensation for the next 18-30 months — and, because each next offer anchors off this one, will materially affect your earnings for the next decade. Adjusted for hourly rate, it is the most valuable hour you spend in your entire career, and almost no engineer treats it that way.

The reason students underprepare for this conversation is psychological, not informational. Gratitude — they offered me a job, I shouldn’t push it. Fear — they’ll rescind if I push back. Imposter feeling — I haven’t earned the right to ask for more. All three are wrong. Recruiters expect negotiation. Offers are almost never rescinded over a polite, well-reasoned counter. The number on the offer letter is the opening, not the conclusion. The candidates who don’t understand that are the ones leaving 10-30% on the table.

This module is the practical version of “negotiate your offer.” Specifically: the script, the data sources, the four components, and the boundary between “should ask” and “should stop pushing.” Take it seriously — the reps are very few and the stakes are very high.

Set up — your data before any conversation

Before you say anything to the recruiter, gather:

• Levels.fyi data for your role and level. P50, P75, P90 numbers. Filter by location, by year, by company stage.
• 2-3 anonymous data points from people you know. Someone at the company, someone at a peer company, someone who left recently. Honest numbers in private.
• Your walk-away. The number below which you’d say no. Write it down. In writing. Negotiation in your head is negotiation lost.
• Your target. The number you’d be very happy with. Above your walk-away, below the moon. Realistic.
• A competing offer if at all possible. This is the single biggest piece of leverage. We’ll talk about how to manufacture one.

The data takes 2-3 hours to gather. Most students skip it and walk in with a number from a Reddit thread. Don’t.

Read these first

Three sources, in this order, then practice:

1. Patrick McKenzie — Salary Negotiation: Make More Money, Be More Valued. post · 60 min · the canonical piece. Read it twice. Most of the script in this module is downstream of it.
2. Haseeb Qureshi — Ten Rules for Negotiating a Job Offer. post · 30 min · the modern bookend, with specific dialogue. Read it once.
3. Josh Doody — Fearless Salary Negotiation (chapters on counter offers and recruiter dynamics). book · 60 min · the long version. Useful if you’re nervous about the actual conversation.

Stop. Don’t read random YouTube negotiation content. Patio11 and Haseeb’s pieces, plus the data sources, are 95% of what you need.

Why students under-negotiate

Three reasons, all worth naming directly:

• Gratitude. “They picked me, I’d feel ungrateful asking for more.” This conflates the relationship with the transaction. The hiring manager who liked you in interviews still likes you after you counter; the recruiter explicitly expects it.
• Fear of rescission. “What if they pull the offer?” This essentially never happens for a polite, professional negotiation. The cases where it does happen are when candidates make impossible demands, lie about competing offers, or are rude. Don’t do those things; you’re fine.
• Imposter feeling. “I haven’t earned the right to ask for more.” The market — the data on levels.fyi — has already decided what your role is worth. You’re not asking for charity. You’re asking to be paid what the role is worth.

Internalising those three before the conversation removes 80% of the difficulty.

The leverage stack

In rough order, from most to least leverage:

1. A competing offer, in writing, with a deadline. This is by far the strongest. Recruiters’ decision-making process changes entirely when there’s a number to beat.
2. A strong walk-away. “I have another opportunity I’m seriously considering and I’d need to see X to choose this.” Even without a written competing offer, a credible alternative (current job, grad school, a startup you’re considering joining) gives you spine.
3. Specific market data. “Levels.fyi shows L3 at peer companies in this market at $185k base.” The recruiter has the same data; you using it makes the conversation a calibration, not a beg.
4. Specific value you’d bring. “I have specific experience with X, which is why this role appealed to me.” Useful as supporting context. Not enough on its own.
5. Nothing. “I would just really appreciate more money.” This is what most students lead with. It moves the number by approximately zero.

The big move that students don’t realise they have: manufacture a competing offer. You apply to 4-6 companies in parallel, time the timeline so offers arrive within a 2-week window, and use each one against the others. This is not unethical; it’s exactly how the market is supposed to work. A recruiter’s first question is often “what’s your timeline?” — that’s their way of asking how to position the offer. Plan for parallel offers from the start.

Reading levels.fyi

Levels.fyi is the closest thing to a public salary database. Some specifics:

• The numbers are self-reported. They skew slightly high (people are more likely to report when they’re proud of their offer). Discount the headline number by ~5% for honesty.
• Filter by year and location. A 2021 SF offer is not a 2026 SF offer. Layoffs and the AI hiring boom both moved the bands.
• P50 / P75 / P90. P50 is the median offer. P75 is what a strong negotiator gets. P90 is a stretch that requires either a competing offer or a strongly differentiated candidate.
• Look at the band, not the point estimate. Companies have bands per level. The L4 band at company X might be $170k-220k base; you’re trying to land in the upper half, not asking for a number outside the band.

For roles outside the standard FAANG-tier, supplement with Glassdoor (less reliable but broader coverage) and the H1B salary database (extremely reliable for companies that sponsor; salaries are filed with the government).

The four components — negotiate together, not separately

Every senior offer has roughly four components. Newer engineers often only see and negotiate base salary, leaving 30%+ on the table.

Component	What it is	Negotiation room
Base salary	The cash you get every two weeks	5-15% room. Anchored by level/band.
Equity	Stock grants, vested over 4 years	Often the largest component. 10-50% room.
Sign-on bonus	One-time cash, usually paid in chunks over year 1	High room — recruiters have discretion here.
Performance bonus	% of base, paid annually based on perf	Usually fixed by level, but sometimes % is negotiable.

Three things to internalise:

• Equity is often the most negotiable. Recruiters have more discretion on stock grants than on base, especially at public companies. “I love the role, the base works for me, but I’d like to see the equity grant come up to X” is a very common, very effective conversation.
• Sign-on is the recruiter’s safety valve. When they can’t move base or equity, they can almost always find sign-on. It’s also a one-time cost to them, which is why it’s flexible.
• Equity is not cash. A $200k/year equity grant at a private startup is not equivalent to $200k of base. Discount aggressively for: dilution, the chance the company never IPOs, the 4-year cliff, and your own risk tolerance. Levels.fyi has a calculator. Use it.

The mistake to avoid: negotiating base only, accepting, and discovering the equity grant was 30% below band. Get all four numbers up front. Negotiate the package, not the salary.

The script — best, good, fallback

The actual conversation. You will be asked, usually by phone or video: “What do you think of the offer?”

Best case — you have a competing offer

“Thanks so much for the offer. I’m excited about the team and the role. I do have an offer from $other_company at $other_number_total_comp, and I’d like to come work with you, but I’d need to see the comp closer to that number to make this decision easy. Specifically — could you take another look at the equity component and the sign-on?”

Specific, polite, actionable. You named a number, named the components you want moved, and signaled you’re willing to choose them.

Good case — strong walk-away, no written offer

“Thanks for the offer. The role really excites me and the team has been great in interviews. Based on the market data I’m seeing for this level — levels.fyi shows comparable roles at peer companies in the $X-$Y range — I was hoping to see total comp closer to $Z. Is there room to take another look at the package, especially the equity?”

The data is doing the work. You’re not making a demand; you’re observing a gap and asking for it to be looked at.

Fallback — no leverage, just asking

“Thanks for the offer. I’m really interested in joining. Before I sign, I wanted to ask — is there any flexibility on the package? I’d be especially grateful for some movement on the sign-on bonus or the equity component. Even a modest bump would help me make this work financially.”

Less leverage, but the script still works. You’re polite, you’re specific about which components, and you’ve made it easy for the recruiter to give you a small win that closes the deal.

After you make the ask: shut up. This is the hardest part. The recruiter will respond with something. Wait. Don’t fill the silence. Don’t apologise. The first person to talk in negotiation often loses.

What to negotiate beyond money

The non-cash items that can be more valuable than salary, depending on your situation:

• Start date. Want a month off before starting? Just ask. Recruiters almost always say yes. This is one of the easiest wins and most students don’t take it.
• Level. “I think this role might actually be a level up from L3, given my experience. Could we discuss leveling?” Moving levels (L3 to L4) is rare but huge — you’re shifting your comp band, not just your number.
• Team. Especially at large companies, the team you join can be more important than the salary. Ask which teams have openings; pick deliberately. Nobody negotiates this and it costs them years.
• Remote / location. “Could the role be remote?” or “Could I be based in $cheaper_city for the same package?” Sometimes yes, especially in 2026’s hybrid market.
• Performance review timing. “Could my first review be at 6 months instead of 12?” If you’re confident you’ll outperform, this accelerates your raise.

Negotiate any of these alongside the comp conversation, not after the package is signed.

When not to negotiate

Two cases where you should accept the offer as written, or close to it:

• You’re getting a stretch offer at a stretch company. A junior offer at the company you’d most want to work for, where the offer is already strong, is not the place to push 15%. The downside (irritated team, bad start) outweighs the upside (a few thousand dollars).
• You’d be picky about a 5% bump on a 50%-life-changing offer. If a startup offers you 30% more than your alternative and a meaningful equity grant, taking it as written is sometimes the right move.

The principle: negotiate when you have leverage, accept when the offer is already in your top 10% of outcomes and pushing risks the relationship.

Common counters and how to respond

They say	You say
”This is our best offer."	"I appreciate that. Before we close this, could you double-check on equity and sign-on with the team? I’d love to make this work."
"We can’t go higher than X on base, but could move on sign-on."	"Thanks. I appreciate that. Could we look at year 2 and 3 of equity as well, since that’s where the real comp lives?"
"We don’t negotiate offers.”	(Politely) “I understand. Even small movement on sign-on or start date would help me close this. Can we discuss any of those?"
"What’s your current salary?"	"I’d rather focus on what the role is worth at your company. Based on market data, I was hoping for $X.” (You don’t have to answer this, in many states they can’t legally ask.)
”What other offers do you have?”	If you have a real offer: name it. If you don’t: “I’m in late stages with a few other companies.” Don’t lie.

The “don’t lie about competing offers” warning

Tempting and dangerous. Recruiters talk to each other. Recruiters call the company you claimed offered you the job. Faking an offer is the one move that genuinely does sometimes get rescinded, and the industry is small — a fabrication discovered 5 years later still ends a relationship.

Truthful framings of partial leverage are fine: “I’m in late stages elsewhere,” “I have another offer in hand at $X,” “I’m seriously considering staying at my current company.” Lies about specific competing offers — don’t.

Debrief — log the offer

After every offer cycle, write down:

• The full package: base, equity grant, sign-on, perf bonus %.
• The level / title.
• What you asked for, what they said, what the final was.
• What worked. What you’d do differently.

This file becomes your most valuable career artifact. By offer #3, you have a personal benchmark — a calibration of what you’re worth at each stage — that nothing on the internet can give you.

Going deeper

When you have specific questions, in this order:

1. Patrick McKenzie’s piece (above), reread, after you’ve actually done one negotiation.
2. Levels.fyi — explore, not just for your role.
3. Chris Voss — Never Split the Difference. book · 4 hrs · written for hostage negotiation, useful for any high-stakes conversation. Skim, don’t memorise.
4. Holloway Guide to Negotiating Salary — comprehensive, paid resource. Worth it for senior offers.

Skip “negotiation hacks” YouTube. The 5-minute version of this skill doesn’t exist.

Checkpoints

If any one wobbles, the corresponding section above is what to reread.

1. Name the three reasons students under-negotiate. Which applies to you?
2. List the leverage stack from highest to lowest. Where do you currently sit, and what could move you up one rung?
3. What are the four offer components? Which is usually the most negotiable, and why?
4. Walk through the negotiation script for the “good case” — strong walk-away, no written offer. What’s the role of market data in this script?
5. When should you not negotiate? Why is the answer not “always negotiate”?

If you can answer all five, you’ve earned 07.4. Next is 07.5 — the longer-arc decision about what to specialise in, when, and why.